ContextLab · paxtonfitzpatrick · Dec 1, 2025 · Nov 24, 2025 · Nov 24, 2025 · Nov 24, 2025
diff --git a/.gitignore b/.gitignore
@@ -37,6 +37,7 @@ exp/server.log
 # latex extensions
 paper/**/*.aux
 paper/**/*.fdb_latexmk
+paper/**/*.fff
 paper/**/*.fls
 paper/**/*.log
 paper/**/*.out
@@ -45,3 +46,6 @@ paper/**/*.blg
 paper/**/*.bbl
 texput.log
 paper/.#compile.sh
+
+# MS Word temp files
+paper/**/~$*
diff --git a/README.md b/README.md
@@ -1,4 +1,4 @@
-# Text embedding models yield high-resolution insights into conceptual knowledge from short multiple-choice quizzes
+# Text embedding models yield detailed conceptual knowledge maps derived from short multiple-choice quizzes
 
 <p align="center">
   <a href="https://psyarxiv.com/dh3q2">
@@ -7,8 +7,8 @@
 </p>
 
 This repository contains all data and code used to produce the paper
-"[_Text embedding models yield high-resolution insights into conceptual
-knowledge from short multiple-choice quizzes_](https://psyarxiv.com/dh3q2)" by
+"[_Text embedding models yield detailed conceptual
+knowledge maps derived from short multiple-choice quizzes_](https://psyarxiv.com/dh3q2)" by
 Paxton C. Fitzpatrick, Andrew C. Heusser, and Jeremy R. Manning.
 
 We also include reproducible environments for running our experiment and
@@ -17,7 +17,7 @@ analyses via [Docker](https://www.docker.com/).
 
 ## Table of Contents
 
-- [Text embedding models yield high-resolution insights into conceptual knowledge from short multiple-choice quizzes](#text-embedding-models-yield-high-resolution-insights-into-conceptual-knowledge-from-short-multiple-choice-quizzes)
+- [Text embedding models yield detailed conceptual knowledge maps derived from short multiple-choice quizzes](#text-embedding-models-yield-detailed-conceptual-knowledge-maps-derived-from-short-multiple-choice-quizzes)
   - [Table of Contents](#table-of-contents)
   - [Repo Organization](#repo-organization)
   - [Installing Docker](#installing-docker)

diff --git a/code/notebooks/main/4_reconstructing-knowledge.ipynb b/code/notebooks/main/4_reconstructing-knowledge.ipynb
diff --git a/code/notebooks/main/6_knowledge-smoothness.ipynb b/code/notebooks/main/6_knowledge-smoothness.ipynb
@@ -121,7 +121,7 @@
     "        A `pandas.Series` whose values are iterables of length 2 \n",
     "        representing lower and upper confidence interval bounds for the \n",
     "        accross-participants mean overall/raw p(correct) for each quiz.\n",
-    "    interp_func : float, optional\n",
+    "    interp_freq : float, optional\n",
     "        If provided, interpolate the series of by-distance p(correct)\n",
     "        values for each quiz & accuracy of reference question to the \n",
     "        given frequency before computing the intersections.\n",
@@ -1296,7 +1296,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.7"
+   "version": "3.9.18"
   }
  },
  "nbformat": 4,

diff --git a/paper/admin/Author_Checklist_NCOMMS-23-10640C__1762787328_45.docx b/paper/admin/Author_Checklist_NCOMMS-23-10640C__1762787328_45.docx
diff --git a/paper/admin/nr-reporting-summary.pdf b/paper/admin/nr-reporting-summary.pdf
diff --git a/paper/figs/active-topics.pdf b/paper/figs/active-topics.pdf
diff --git a/paper/figs/bos-qcorrs-peaks.pdf b/paper/figs/bos-qcorrs-peaks.pdf
diff --git a/paper/figs/content-mastery.pdf b/paper/figs/content-mastery.pdf
diff --git a/paper/figs/experiment.pdf b/paper/figs/experiment.pdf
diff --git a/paper/figs/forces-qcorrs-peaks.pdf b/paper/figs/forces-qcorrs-peaks.pdf
diff --git a/paper/figs/individual-knowledge-maps-quiz1.pdf b/paper/figs/individual-knowledge-maps-quiz1.pdf
diff --git a/paper/figs/individual-knowledge-maps-quiz2.pdf b/paper/figs/individual-knowledge-maps-quiz2.pdf
diff --git a/paper/figs/individual-knowledge-maps-quiz3.pdf b/paper/figs/individual-knowledge-maps-quiz3.pdf
diff --git a/paper/figs/individual-learnings-maps-quiz1-2.pdf b/paper/figs/individual-learnings-maps-quiz1-2.pdf
diff --git a/paper/figs/individual-learnings-maps-quiz2-3.pdf b/paper/figs/individual-learnings-maps-quiz2-3.pdf
diff --git a/paper/figs/knowledge-smoothness.pdf b/paper/figs/knowledge-smoothness.pdf
diff --git a/paper/figs/knowledge_and_learning_maps.pdf b/paper/figs/knowledge_and_learning_maps.pdf
diff --git a/paper/figs/lecture-question-similarity.pdf b/paper/figs/lecture-question-similarity.pdf
diff --git a/paper/figs/model-comparison.pdf b/paper/figs/model-comparison.pdf
diff --git a/paper/figs/predict-knowledge-questions.pdf b/paper/figs/predict-knowledge-questions.pdf
diff --git a/paper/figs/sliding_windows.pdf b/paper/figs/sliding_windows.pdf
diff --git a/paper/figs/source/bos-qcorrs-peaks.pdf b/paper/figs/source/bos-qcorrs-peaks.pdf
diff --git a/paper/figs/source/content-mastery.pdf b/paper/figs/source/content-mastery.pdf
diff --git a/paper/figs/source/forces-qcorrs-peaks.pdf b/paper/figs/source/forces-qcorrs-peaks.pdf
diff --git a/paper/figs/source/individual-knowledge-maps-quiz1.pdf b/paper/figs/source/individual-knowledge-maps-quiz1.pdf
diff --git a/paper/figs/source/individual-knowledge-maps-quiz2.pdf b/paper/figs/source/individual-knowledge-maps-quiz2.pdf
diff --git a/paper/figs/source/individual-knowledge-maps-quiz3.pdf b/paper/figs/source/individual-knowledge-maps-quiz3.pdf
diff --git a/paper/figs/source/individual-learnings-maps-quiz1-2.pdf b/paper/figs/source/individual-learnings-maps-quiz1-2.pdf
diff --git a/paper/figs/source/individual-learnings-maps-quiz2-3.pdf b/paper/figs/source/individual-learnings-maps-quiz2-3.pdf
diff --git a/paper/figs/source/lecture-question-similarity.pdf b/paper/figs/source/lecture-question-similarity.pdf
diff --git a/paper/figs/source/topic-weights-corrmat.pdf b/paper/figs/source/topic-weights-corrmat.pdf
diff --git a/paper/figs/source/topic-weights.pdf b/paper/figs/source/topic-weights.pdf
diff --git a/paper/figs/topic-weights.pdf b/paper/figs/topic-weights.pdf
diff --git a/paper/figs/word-overlap-comparison.pdf b/paper/figs/word-overlap-comparison.pdf
diff --git a/paper/main.pdf b/paper/main.pdf
diff --git a/paper/main.tex b/paper/main.tex
diff --git a/paper/supplement.pdf b/paper/supplement.pdf
diff --git a/paper/supplement.tex b/paper/supplement.tex
@@ -32,8 +32,8 @@
 \newcommand{\knowledgeMaps}{7}
 \newcommand{\topicModelMethods}{\textit{Constructing text embeddings of multiple lectures and questions}}
 
-\title{\textit{Supplementary materials for}: Text embedding models yield 
-high-resolution insights into conceptual knowledge from short multiple-choice
+\title{\textit{Supplementary information for}: Text embedding models yield 
+detailed conceptual knowledge maps derived from short multiple-choice
 quizzes}
 
 \author{Paxton C. Fitzpatrick\textsuperscript{1}, Andrew C. 
@@ -326,7 +326,7 @@ \section*{Supplementary figures}
     \caption{\textbf{Individual participants' knowledge maps estimated from
     Quiz 1 responses.} Each panel is in the same format as the knowledge map
     displayed in the left panel of Figure~\knowledgeMaps A in the main text,
-    but here the maps are shown separately for each participant.}
+    but here each participant's map shows only the questions they answered on Quiz 1.}
 
     \label{fig:knowledge-maps-q1}
 \end{figure}
@@ -339,7 +339,7 @@ \section*{Supplementary figures}
     \caption{\textbf{Individual participants' knowledge maps estimated from
     Quiz 2 responses.} Each panel is in the same format as the knowledge map
     displayed in the middle panel of Figure~\knowledgeMaps A in the main text,
-    but here the maps are shown separately for each participant.}
+    but here each participant's map shows only the questions they answered on Quiz 2.}
 
     \label{fig:knowledge-maps-q2}
 \end{figure}
@@ -352,7 +352,7 @@ \section*{Supplementary figures}
     \caption{\textbf{Individual participants' knowledge maps estimated from
     Quiz 3 responses.} Each panel is in the same format as the knowledge map
     displayed in the right panel of Figure~\knowledgeMaps A in the main text,
-    but here the maps are shown separately for each participant.}
+    but here each participant's map shows only the questions they answered on Quiz 3.}
 
     \label{fig:knowledge-maps-q3}
 \end{figure}
@@ -365,7 +365,7 @@ \section*{Supplementary figures}
     \caption{\textbf{Individual participants' learning maps estimated from
     Quiz 1 and 2 responses.} Each panel is in the same format as the learning map
     displayed in the left panel of Figure~\knowledgeMaps B in the main text,
-    but here the maps are shown separately for each participant.}
+    but here each participant's map shows only the questions they answered on Quizzes 1 \& 2.}
 
     \label{fig:learning-maps-q1_2}
 \end{figure}
@@ -378,7 +378,7 @@ \section*{Supplementary figures}
     \caption{\textbf{Individual participants' learning maps estimated from
     Quiz 2 and 3 responses.} Each panel is in the same format as the learning map
     displayed in the right panel of Figure~\knowledgeMaps B in the main text,
-    but here the maps have been computed separately for each participant.}
+    but here each participant's map shows only the questions they answered on Quizzes 2 \& 3.}
 
     \label{fig:learning-maps-q2_3}
 \end{figure}
@@ -400,7 +400,7 @@ \section*{Supplementary figures}
 
 \begin{figure}[tp]
     \centering
-    \includegraphics[width=0.57\textwidth]{figs/word-overlap-comparison}
+    \includegraphics[width=0.52\textwidth]{figs/word-overlap-comparison}
 
     \caption{\textbf{Topic correlations versus word overlap.} \textbf{A.
     Single-timepoint comparisons for \textit{Four Fundamental Forces}.} Each
@@ -428,7 +428,7 @@ \section*{Supplementary figures}
 
 \FloatBarrier
 
-\section*{Supplementary results}
+\section*{Supplementary discussion}
 
 \doublespacing